All BCM63XX SoCs starting with BCM6358 have a BMIPS4350 instead of a
BMIPS3300, so select it unless support for any of the older SoCs is
selected.
All BMIPS4350 have only two CPUs, so select the appropriate default.
Signed-off-by: NJonas Gorski <jogo@openwrt.org>
Cc: linux-mips@linux-mips.org
Cc: John Crispin <blogic@openwrt.org>
Cc: Maxime Bizon <mbizon@freebox.fr>
Cc: Florian Fainelli <florian@openwrt.org>
Cc: Kevin Cernekee <cernekee@gmail.com>
Patchwork: https://patchwork.linux-mips.org/patch/5355/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

MIPS software IRQs 0 and 1 are used for interprocessor signaling (IPI)
on BMIPS SMP.  Make the board support code aware of them.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
[jogo@openwrt.org: move sw irqs behind timer irq]
Signed-off-by: NJonas Gorski <jogo@openwrt.org>
Cc: linux-mips@linux-mips.org
Cc: John Crispin <blogic@openwrt.org>
Cc: Maxime Bizon <mbizon@freebox.fr>
Cc: Florian Fainelli <florian@openwrt.org>
Patchwork: https://patchwork.linux-mips.org/patch/5354/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

For non-SMP, uses the new random canary value that is stored in the
task struct whenever a new task is forked.  Based on ARM version in
df0698be and subject to the same
limitations: the variable GCC expects, __stack_chk_guard, is global,
so this will not work on SMP.

Quoting Nicolas Pitre <nico@fluxnic.net>: "One way to overcome this
GCC limitation would be to locate the __stack_chk_guard variable into
a memory page of its own for each CPU, and then use TLB locking to
have each CPU see its own page at the same virtual address for each of
them."
Signed-off-by: NGregory Fong <gregory.0xf0@gmail.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5488/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Implements basic stack protector support based on ARM version in
c743f380 , with Kconfig option,
constant canary value set at boot time, and script to check if
compiler actually supports stack protector.

Tested by creating a kernel module that writes past end of char[].
Signed-off-by: NGregory Fong <gregory.0xf0@gmail.com>
Cc: linux-mips@linux-mips.org
Cc: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Cc: Carmelo Amoroso <carmelo.amoroso@st.com>
Patchwork: https://patchwork.linux-mips.org/patch/5448/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

When building with -fstack-protector, gcc emits the __stack_chk_guard and
__stack_chk_fail symbols to check for stack stability. These symbols are
defined in vmlinux but the generated vmlinux.bin that is used to create
the compressed vmlinuz image has no symbol table so the linker can't find
these symbols during the final linking phase.  As a result of which, we
need either to redefine these symbols just for the compressed image or drop
the -fstack-protector option when building the compressed image. This patch
implements the latter of two options.

Fixes the following linking problem:

dbg.c:(.text+0x7c): undefined reference to `__stack_chk_guard'
dbg.c:(.text+0x80): undefined reference to `__stack_chk_guard'
dbg.c:(.text+0xd4): undefined reference to `__stack_chk_guard'
dbg.c:(.text+0xec): undefined reference to `__stack_chk_fail'

[ralf@linux-mips.org: I'm applying this before the patch that actually adds
stack protector support for MIPS.  This means, it will not be possible
to trigger above error message with any commit from the tree but rather
they are what one would hit without this commit.]
Signed-off-by: NMarkos Chandras <markos.chandras@imgtec.com>
Cc: linux-mips@linux-mips.org
Cc: Markos Chandras <markos.chandras@imgtec.com>
Patchwork: https://patchwork.linux-mips.org/patch/5575/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The MIPS interrupt controllers are all registering their own irq_domains
now. Drop the MIPS specific code because it is no longer needed.
Signed-off-by: NGrant Likely <grant.likely@linaro.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/5458/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This typedef is unnecessary and should just be removed.
Signed-off-by: NJoe Perches <joe@perches.com>
Cc: Jiri Kosina <trivial@kernel.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/5460/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The PMC MSP71XX gpio drivers were added in v2.6.28, see commit
9fa32c6b ("MIPS: PMC MSP71XX gpio drivers"). They are only built if
CONFIG_HAVE_GPIO_LIB is set.

But the Kconfig symbol HAVE_GPIO_LIB was already removed in v2.6.27, see
commit 7444a72e ("gpiolib: allow user-selection"). So these drivers
were never buildable. Perhaps no-one noticed because there are no in
tree users of msp71xx_init_gpio() and msp71xx_init_gpio_extended().
Anyhow, these drivers can safely be removed.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/5345/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

promcon.o is built if CONFIG_PROM_CONSOLE is set. But there's no Kconfig
symbol PROM_CONSOLE, so promcon.c is unbuildable. Remove it.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Acked-by: NMaciej W. Rozycki <macro@linux-mips.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/5344/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The invalid value for scratch register is -1, so update the checks of
the form (scratch_reg > 0) to be (scratch_reg >= 0).  This will fix
the case in Netlogic XLP where the scratch_reg can be 0.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5444/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

XLR/XLP COP0 scratch is register 22, sel 0-7. Add a function
c0_kscratch() which returns the scratch register for the platform,
and use the return value while generating TLB handlers.

Setup kscratch_mask to 0xf for XLR/XLP since the config4 register
does not exist. This allows the kernel to allocate scratch registers
0-3 if needed.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5445/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix an issue in plat_irq_dispatch due to which it can call do_IRQ
with a PIC irq that is not mapped.

When a per-cpu interrupt and a PIC interrupt are both active, the
check 'eirr & PERCPU_IRQ_MASK' will be true, but the interrupt in 'i'
will be the number of the PIC interrupt. In this case, we will call
do_IRQ on the PIC interrupt without mapping it with nlm_irq_to_xirq().

Fix this by using __ffs64 instead of __ilog2_u64 and using the
interrupt number instead of mask to identify per-cpu interrupts.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5432/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The XLPs in production do not need these workarounds. Remove the code and
the associated ifdef.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5430/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix a cache error found in stress test, caused by the prefetch instruction
going beyond valid memory when acessing the last page of a region. Add
the pref_backup logic similar to XLR in XLP too.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5431/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Earlier we just waited for the first thread of the CPU to come online
before proceeding to wake up others. Update it to wait for all the CPUs
in the core. This will be useful when the boot-up is slow, like while
debugging or when running in a simulator.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5429/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This does not cause a problem yet, but we do not want to write 1
to reserved bits.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5424/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix an issue in the reset code. Since this code is copied to the
reset vector, using 'j' for looping is not correct. Use relative
branch 'b'.

Update the usage of 'j' in smpboot.S to be consistent although it
is not a bug there.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5427/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Move the nlm_cpu_ready[] array used by the cpu wakeup code to the
boot area, along with rest of the boot parameter code.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5425/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This moves the calculation and casting needed to access the CPU initialization
data to a function nlm_get_boot_data()
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5426/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The core initialization and reset vector setup needs to be done
even when booting uniprocessor. Move this code from smp.c to setup.c
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5428/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The reset and core initialization code should be available for
uniprocessor as well. This changes is just to take out the code
into a different file, without any change to the logic.

The change for  uniprocessor initialization code is in a later patch.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5423/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Create new flle arch/mips/netlogic/xlp/dt.c and move the device
tree related code there.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5422/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add SYS_SUPPORTS_ZBOOT and SYS_SUPPORTS_ZBOOT_UART16550 config options
for XLR and XLP.

Update boot/compressed/uart-16550.c to add UART port for XLR and XLP.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5417/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix uart-16550.c for adding XLR/XLP support, changes are:
* Make register read/write use volatile pointers
* Support 32 bit IO read/write
* Increase timeout in waiting for UART LSR
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5416/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add COP2 register state structure and functions for Netlogic XLP. The
RX and TX buffers and status registers are to be saved. Since the
registers are 64-bit, do the implementation in inline assembly which
works on both 32-bit and 64-bit kernels.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: ddaney.cavm@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5413/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Rename macro nlm_cop2_enable() to nlm_cop2_enable_irqsave() and the macro
nlm_cop2_restore to nlm_cop2_disable_irqrestore(). The new names will
reflect the functionality better, and will make nlm_cop2_restore()
available to be used later in COP2 save/restore patch.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: ddaney.cavm@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5412/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The sel argument os nlm_read_c2_status() was not used and the macro
returned the sel 0 in all cases. Fix this by defining two macros:
nlm_read_c2_status0() and nlm_read_c2_status1() to return the two
status registers.

Add functions to write to the status registers as well.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: ddaney.cavm@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5414/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Kernel threads should be able to use COP2 if the platform needs it.
Do not call die_if_kernel() for a coprocessor unusable exception if
the exception due to COP2 usage.  Instead, the default notifier for
COP2 exceptions is updated to call die_if_kernel.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: ddaney.cavm@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5415/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Move the common code for saving and restoring platform specific COP2
registers to switch_to(). This will make supporting new platforms (like
Netlogic XLP) easier.

The platform specific COP2 definitions are to be specified in
asm/processor.h and in asm/cop2.h.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: ddaney.cavm@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5411/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add SWIOTLB config option and related files to Netlogic platform.

Some XLP SoC components like the SD/MMC interface cannot do DMA beyond
32-bit physical address. The SD/MMC driver can use memory outside this
range for IO, to support this we have to add bounce buffers implemented
by SWIOTLB.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: Ganesan Ramalingam <ganesanr@broadcom.com>
Patchwork: https://patchwork.linux-mips.org/patch/5410/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Provide a default implementation of phys_to_dma and dma_to_phys in
mach-generic/dma_coherence.h.

If CONFIG_NEED_SG_DMA_LENGTH is defined, the dma_length field in
struct scatterlist is used. Set this up in mips_dma_map_sg so that
the default mips DMA ops can be used when SWIOTLB is enabled.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5409/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

CONFIG_FRAME_POINTER is not selectable for MIPS so this
codepath was never executed.
Signed-off-by: NMarkos Chandras <markos.chandras@imgtec.com>
Acked-by: NSteven J. Hill <Steven.Hill@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5440/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This enables support for CONFIG_NO_HZ_FULL.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This will simplify further modifications.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Ensure strings are always '\0' terminated.  Or in the next pr_info() shit
may hit the fan.
Signed-off-by: NChen Gang <gang.chen@asianux.com>
Acked-by: NJohn Crispin <blogic@openwrt.org>
Cc: david.daney@cavium.com
Cc: linux-mips@linux-mips.org
Cc: Linux-Arch <linux-arch@vger.kernel.org>
Cc: linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org>
Patchwork: https://patchwork.linux-mips.org/patch/5331/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

'compatible' is used by strlen() in __of_device_is_compatible().

Ensure strings are always '\0' terminated.

'of_ids is not a structure in "include/uapi/*", so no need to initialize
it completly; using strlcpy() instead of strncpy() will do.
Signed-off-by: NChen Gang <gang.chen@asianux.com>
Acked-by: NJohn Crispin <blogic@openwrt.org>
Cc: juhosg@openwrt.org
Cc: linux-mips@linux-mips.org
Cc: Linux-Arch <linux-arch@vger.kernel.org>
Patchwork: https://patchwork.linux-mips.org/patch/5330/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

'compatible' is used by strlen() in __of_device_is_compatible().

Ensure strings are always '\0' terminated.

'of_ids is not a structure in "include/uapi/*", so no need to initialize
it completly; using strlcpy() instead of strncpy() will do.
Signed-off-by: NChen Gang <gang.chen@asianux.com>
Acked-by: NJohn Crispin <blogic@openwrt.org>
Cc: linux-mips@linux-mips.org
Cc: Linux-Arch <linux-arch@vger.kernel.org>
Patchwork: https://patchwork.linux-mips.org/patch/5329/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

When you turn it off, the kernel is unusable, so get rid of the option
and always allow unaligned access.

The Octeon specific memcpy intentionally does unaligned accesses and it
must not fault.
Signed-off-by: NDavid Daney <david.daney@cavium.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5303/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>